youtube_dl/extractor/udemy.py

   1 from __future__ import unicode_literals
   2
   3 from .common import InfoExtractor
   4 from ..compat import (
   5     compat_HTTPError,
   6     compat_urllib_parse,
   7     compat_urllib_request,
   8 )
   9 from ..utils import (
  10     ExtractorError,
  11     float_or_none,
  12     int_or_none,
  13     sanitized_Request,
  14 )
  15
  16
  17 class UdemyIE(InfoExtractor):
  18     IE_NAME = 'udemy'
  19     _VALID_URL = r'https?://www\.udemy\.com/(?:[^#]+#/lecture/|lecture/view/?\?lectureId=)(?P<id>\d+)'
  20     _LOGIN_URL = 'https://www.udemy.com/join/login-popup/?displayType=ajax&showSkipButton=1'
  21     _ORIGIN_URL = 'https://www.udemy.com'
  22     _SUCCESSFULLY_ENROLLED = '>You have enrolled in this course!<'
  23     _ALREADY_ENROLLED = '>You are already taking this course.<'
  24     _NETRC_MACHINE = 'udemy'
  25
  26     _TESTS = [{
  27         'url': 'https://www.udemy.com/java-tutorial/#/lecture/172757',
  28         'md5': '98eda5b657e752cf945d8445e261b5c5',
  29         'info_dict': {
  30             'id': '160614',
  31             'ext': 'mp4',
  32             'title': 'Introduction and Installation',
  33             'description': 'md5:c0d51f6f21ef4ec65f091055a5eef876',
  34             'duration': 579.29,
  35         },
  36         'skip': 'Requires udemy account credentials',
  37     }]
  38
  39     def _enroll_course(self, webpage, course_id):
  40         enroll_url = self._search_regex(
  41             r'href=(["\'])(?P<url>https?://(?:www\.)?udemy\.com/course/subscribe/.+?)\1',
  42             webpage, 'enroll url', group='url',
  43             default='https://www.udemy.com/course/subscribe/?courseId=%s' % course_id)
  44         webpage = self._download_webpage(enroll_url, course_id, 'Enrolling in the course')
  45         if self._SUCCESSFULLY_ENROLLED in webpage:
  46             self.to_screen('%s: Successfully enrolled in' % course_id)
  47         elif self._ALREADY_ENROLLED in webpage:
  48             self.to_screen('%s: Already enrolled in' % course_id)
  49
  50     def _download_lecture(self, course_id, lecture_id):
  51         return self._download_json(
  52             'https://www.udemy.com/api-2.0/users/me/subscribed-courses/%s/lectures/%s?%s' % (
  53                 course_id, lecture_id, compat_urllib_parse.urlencode({
  54                     'video_only': '',
  55                     'auto_play': '',
  56                     'fields[lecture]': 'title,description,asset',
  57                     'fields[asset]': 'asset_type,stream_url,thumbnail_url,download_urls,data',
  58                     'instructorPreviewMode': 'False',
  59                 })),
  60             lecture_id, 'Downloading lecture JSON')
  61
  62     def _handle_error(self, response):
  63         if not isinstance(response, dict):
  64             return
  65         error = response.get('error')
  66         if error:
  67             error_str = 'Udemy returned error #%s: %s' % (error.get('code'), error.get('message'))
  68             error_data = error.get('data')
  69             if error_data:
  70                 error_str += ' - %s' % error_data.get('formErrors')
  71             raise ExtractorError(error_str, expected=True)
  72
  73     def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata'):
  74         headers = {
  75             'X-Udemy-Snail-Case': 'true',
  76             'X-Requested-With': 'XMLHttpRequest',
  77         }
  78         for cookie in self._downloader.cookiejar:
  79             if cookie.name == 'client_id':
  80                 headers['X-Udemy-Client-Id'] = cookie.value
  81             elif cookie.name == 'access_token':
  82                 headers['X-Udemy-Bearer-Token'] = cookie.value
  83                 headers['X-Udemy-Authorization'] = 'Bearer %s' % cookie.value
  84
  85         if isinstance(url_or_request, compat_urllib_request.Request):
  86             for header, value in headers.items():
  87                 url_or_request.add_header(header, value)
  88         else:
  89             url_or_request = sanitized_Request(url_or_request, headers=headers)
  90
  91         response = super(UdemyIE, self)._download_json(url_or_request, video_id, note)
  92         self._handle_error(response)
  93         return response
  94
  95     def _real_initialize(self):
  96         self._login()
  97
  98     def _login(self):
  99         (username, password) = self._get_login_info()
 100         if username is None:
 101             return
 102
 103         login_popup = self._download_webpage(
 104             self._LOGIN_URL, None, 'Downloading login popup')
 105
 106         def is_logged(webpage):
 107             return any(p in webpage for p in ['href="https://www.udemy.com/user/logout/', '>Logout<'])
 108
 109         # already logged in
 110         if is_logged(login_popup):
 111             return
 112
 113         login_form = self._form_hidden_inputs('login-form', login_popup)
 114
 115         login_form.update({
 116             'email': username.encode('utf-8'),
 117             'password': password.encode('utf-8'),
 118         })
 119
 120         request = sanitized_Request(
 121             self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
 122         request.add_header('Referer', self._ORIGIN_URL)
 123         request.add_header('Origin', self._ORIGIN_URL)
 124
 125         response = self._download_webpage(
 126             request, None, 'Logging in as %s' % username)
 127
 128         if not is_logged(response):
 129             error = self._html_search_regex(
 130                 r'(?s)<div[^>]+class="form-errors[^"]*">(.+?)</div>',
 131                 response, 'error message', default=None)
 132             if error:
 133                 raise ExtractorError('Unable to login: %s' % error, expected=True)
 134             raise ExtractorError('Unable to log in')
 135
 136     def _real_extract(self, url):
 137         lecture_id = self._match_id(url)
 138
 139         webpage = self._download_webpage(url, lecture_id)
 140
 141         course_id = self._search_regex(
 142             r'data-course-id=["\'](\d+)', webpage, 'course id')
 143
 144         try:
 145             lecture = self._download_lecture(course_id, lecture_id)
 146         except ExtractorError as e:
 147             # Error could possibly mean we are not enrolled in the course
 148             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
 149                 self._enroll_course(webpage, course_id)
 150                 lecture_id = self._download_lecture(course_id, lecture_id)
 151             else:
 152                 raise
 153
 154         title = lecture['title']
 155         description = lecture.get('description')
 156
 157         asset = lecture['asset']
 158
 159         asset_type = asset.get('assetType') or asset.get('asset_type')
 160         if asset_type != 'Video':
 161             raise ExtractorError(
 162                 'Lecture %s is not a video' % lecture_id, expected=True)
 163
 164         stream_url = asset.get('streamUrl') or asset.get('stream_url')
 165         if stream_url:
 166             youtube_url = self._search_regex(
 167                 r'(https?://www\.youtube\.com/watch\?v=.*)', stream_url, 'youtube URL', default=None)
 168             if youtube_url:
 169                 return self.url_result(youtube_url, 'Youtube')
 170
 171         video_id = asset['id']
 172         thumbnail = asset.get('thumbnailUrl') or asset.get('thumbnail_url')
 173         duration = float_or_none(asset.get('data', {}).get('duration'))
 174         outputs = asset.get('data', {}).get('outputs', {})
 175
 176         formats = []
 177         for format_ in asset.get('download_urls', {}).get('Video', []):
 178             video_url = format_.get('file')
 179             if not video_url:
 180                 continue
 181             format_id = format_.get('label')
 182             f = {
 183                 'url': format_['file'],
 184                 'height': int_or_none(format_id),
 185             }
 186             if format_id:
 187                 # Some videos contain additional metadata (e.g.
 188                 # https://www.udemy.com/ios9-swift/learn/#/lecture/3383208)
 189                 output = outputs.get(format_id)
 190                 if isinstance(output, dict):
 191                     f.update({
 192                         'format_id': '%sp' % (output.get('label') or format_id),
 193                         'width': int_or_none(output.get('width')),
 194                         'height': int_or_none(output.get('height')),
 195                         'vbr': int_or_none(output.get('video_bitrate_in_kbps')),
 196                         'vcodec': output.get('video_codec'),
 197                         'fps': int_or_none(output.get('frame_rate')),
 198                         'abr': int_or_none(output.get('audio_bitrate_in_kbps')),
 199                         'acodec': output.get('audio_codec'),
 200                         'asr': int_or_none(output.get('audio_sample_rate')),
 201                         'tbr': int_or_none(output.get('total_bitrate_in_kbps')),
 202                         'filesize': int_or_none(output.get('file_size_in_bytes')),
 203                     })
 204                 else:
 205                     f['format_id'] = '%sp' % format_id
 206             formats.append(f)
 207
 208         self._sort_formats(formats)
 209
 210         return {
 211             'id': video_id,
 212             'title': title,
 213             'description': description,
 214             'thumbnail': thumbnail,
 215             'duration': duration,
 216             'formats': formats
 217         }
 218
 219
 220 class UdemyCourseIE(UdemyIE):
 221     IE_NAME = 'udemy:course'
 222     _VALID_URL = r'https?://www\.udemy\.com/(?P<id>[\da-z-]+)'
 223     _TESTS = []
 224
 225     @classmethod
 226     def suitable(cls, url):
 227         return False if UdemyIE.suitable(url) else super(UdemyCourseIE, cls).suitable(url)
 228
 229     def _real_extract(self, url):
 230         course_path = self._match_id(url)
 231
 232         webpage = self._download_webpage(url, course_path)
 233
 234         response = self._download_json(
 235             'https://www.udemy.com/api-1.1/courses/%s' % course_path,
 236             course_path, 'Downloading course JSON')
 237
 238         course_id = response['id']
 239         course_title = response.get('title')
 240
 241         self._enroll_course(webpage, course_id)
 242
 243         response = self._download_json(
 244             'https://www.udemy.com/api-1.1/courses/%s/curriculum' % course_id,
 245             course_id, 'Downloading course curriculum')
 246
 247         entries = [
 248             self.url_result(
 249                 'https://www.udemy.com/%s/#/lecture/%s' % (course_path, asset['id']), 'Udemy')
 250             for asset in response if asset.get('assetType') or asset.get('asset_type') == 'Video'
 251         ]
 252
 253         return self.playlist_result(entries, course_id, course_title)